MW lineup

[Mike Ossipoff]

If we define MW in terms of actual votes, as I did in one part of
the previous message, then every method that's been proposed to
this list meets that criterion, though I can't say that I've checked
every one of them out.

If we define MW in the more usual way, by saying that an alternative
that's favorite of a majority should always win, then no method
meets it, with mathematical certainty, because there's always the
possibility that some voters could vote a less-liked alternative
equal to or over a more-liked one.

In Condorcet's method that would be an irrational thing to do.
There'd be no reason to do so. In Regular Champion, or any other
Copeland version, there would often be strategic necessity to
do so. Of course the problem in sw methods, other than the best
ones, is that sometimes we don't know whether we need the compromise
or whether our favorite has a win. Show me a method where that's
a real problem, & I'll show you a method that can't be considered
the best to propose.

If a method is said to fail MW if an alternative that's favorite
of a majority loses, even if voters vote their favorite over
everything else unless there's strategic need not to, then 
Smith//Condorcet & plan Condorcet don't fail MW, but 
Regular-Champion fails it miserably, as do all versions
of Copeland.


Mike











--